linux-kernel - Re: [PATCHv4 0/7] Serial port generic PM to fix 8250 PM

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YaX2mbUv9Yv3icl4@hovoldconsulting.com>
Date:   Tue, 30 Nov 2021 11:02:01 +0100
From:   Johan Hovold <johan@...nel.org>
To:     Tony Lindgren <tony@...mide.com>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Andy Shevchenko <andriy.shevchenko@...el.com>,
        Jiri Slaby <jirislaby@...nel.org>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Vignesh Raghavendra <vigneshr@...com>,
        linux-serial@...r.kernel.org, linux-omap@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Subject: Re: [PATCHv4 0/7] Serial port generic PM to fix 8250 PM

On Mon, Nov 15, 2021 at 10:41:56AM +0200, Tony Lindgren wrote:
> Hi,
> 
> Here are v4 patches for serial port generic PM. The scope has now expanded
> a bit from the earlier attempts to get rid of pm_runtime_irq_safe() for
> the 8250_omap driver. I've now picked up three patches from Andy's earlier
> generic serial port PM series.

So this looks like another step in the right direction but there are
still some missing pieces.

First, you need to provide an overview of the design decisions made here
in cover letter. It's currently spread out over several patches and
those commit messages still do not hold all the details.

Specifically, it looks like tx can still stall indefinitely if the
autosuspend timer fires. This can happen at low baud rates and also when
using flow control.

It also looks like the expected calls to update the last busy timestamp
might be missing from the interrupt handlers or related helpers.

Please also describe how this interacts with the console. Is a console
port now never suspended? Where is that enforced? The final patch
appears to rely on this when it drops PM calls from for example some
console poll callbacks.

> Changes since v3:
> - Pick three patches from Andy's earlier serial port PM series to handle
>   issues pointed out by Johan

Please also be more specific here when sending an updated series. I
can't really tell what has changed from just this one sentence.

Johan